Implementation Of An Automated Flexible Manufacturing System Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Automation is the use of control systems and information technologies reducing the need for human intervention. A fully automated factory needs to have three essential attributes


High level of productivity

Quality products.

An automated system should have high level of productivity, good quality products and a very rapid capability to change the products. All these led to the development of automated flexible manufacturing systems (FMS).

A work piece commonly used in automotive engines is shown in Fig. 1.

Fig. 1: Work piece to be manufactured

It is now desired to mass-produce this work piece in a Flexible Manufacturing System (FMS) framework. The automated manufacturing system should posses the following features

Should able to identify and report if some other than the given raw work piece is loaded in the system

Should detect minute surface and near-surface flaws and abrasion and separate as scrap

Feedbacks from every individual station to check for smooth processing

Deadlock Detection while handling the work piece

Tool wear detection and fault diagnosis

Centralized control unit for continuous monitoring and annunciation

1.1 Various Operations Involved

Turning & Boring:

Turning is the process of machining external cylindrical and conical surfaces. The process uses a machine tool called a lathe. Boring is the process of enlarging an existing hole or internal cylindrical surface. This can be accomplished on a lathe or a machine tool specifically designed for the process such as a horizontal boring machine.

Thread cutting and Shaping

Threading is the process of creating a screw thread. Shaping is the machining operation for generating flat surface by means of single point cutting tool. The work piece is kept stationary and the cutting tool given a reciprocating motion

The FMS structure consists of

Three robotic arms (R1, R2, and R3),

Two flexible machines (M1 and M2), and

Four conveyor belts (B1, B2, B3, and B4) as shown in Fig. 2

Fig. 2: FMS consisting of robotic arms, flexible machines, and conveyor belts


Let the time taken for the work piece to travel end to end in the conveyor belt be T

Time taken by the machine to perform single machining is T

Consider the time taken by the robotic arm for pick and place operation is small and negligible compared to T

the machine M1 performs machining 1 and 3 and M2 performs machining 2 and 4 respectively

1.2 Sequence of operations:

Start process

The raw work piece is first picked up by the robotic arm R1and placed in the conveyor B1.

Machining 1

Once it reaches the end of B1, arm R2 picks it up and places in M1 to undergo machining 1.

Machining 2

After the first operation, it is forwarded to M2 through R2 and B2

Machining 3

Again the same work piece is place in conveyor B3 to reach the machine M1 for machining 3. During this stage R2 should be managed well to avoid ambiguity and the task should be clear

Machining 4

Once again the work piece sent to M2 to perform machining 4 through R2 and B2

End Process

After the completion of four processes R3 places the finished work piece in B4. Robotic arm R1 fetches and stacks at the output

When the work piece is loaded in a conveyor at a given instant of time it shall reach the end of the conveyor belt at Tth time. So the time duration spent by the work piece from the initial time t is (t-T).

Let the

Initial time at the start of process't'

Time taken by the conveyor 'T'

Duration of time to travel is (t to T)

Also to perform a single operation the time taken by the machine is T.

Let's see the operation cycle to determine at which stage there is no ambiguity. It will be easy to sequence the FMS for different operation once the collision stages are determined.

The below logic represents the simple sequence by which the system works

Loading 1st work piece:

Time: t-T T-2T 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R1 R2 R2 R3 R3 R2 R2 R3 R3 R1

U1 B1 M1 B2 M2 B3 M1 B2 M2 B4 Y1

Loading 2nd work piece:

Time: T-2T 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T 9T-10T

R1 R2 R2 R3 R3 R2 R2 R3 R3 R1

U1 B1 M1 B2 M2 B3 M1 B2 M2 B4 Y1

Note that after the completion of time 2T, 1st work piece will be waiting at Machine M1 to pick and place at B2 at the same time 2nd work piece will be waiting for arm R2 to be placed at M1.

At this stage arm R2 will be in trouble, i.e. ambiguity arises.

1.3 Solution to avoid ambiguity

To avoid this lets start the next new cycle at 2T, which provides sufficient time for the robotic arm and there won't be any deadlock or ambiguity.

Loading 1st work piece

Time: t-T T-2T 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R1 R2 R2 R3 R3 R2 R2 R3 R3 R1

U1 B1 M1 B2 M2 B3 M1 B2 M2 B4 Y1

Loading 2nd work piece

Time: 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R2 R2 R3 R3 R2 R2

U1 B1 M1 B2 M2 B3 M1 B2







t to T

T to 2T

Machining 1

2T to 3T

3T to 4T

Machining 1

Machining 2

4T to 5T

5T to 6T

Machining 3

Machining 2

6T to 7T

7T to 8T

Machining 3

Machining 4

8T to 9T

Tab 1.Tabulation to illustrate the consecutive machining operation at different time intervals

First work piece loaded at 't'

Second work piece loaded at't+2T'

Implementation Of Measurement System

The manufacturing system consists of CNC, robotic arms, conveyor belts and sensors. The total system should be automated using adequate sensors and controller.

The entire system is sub divided for easy analysis and troubleshooting. The major systems are classified into following

Conveyor monitoring system

-Belt load receptor

-Speed transmitter

Inductive proximity sensor system

Robotic arm

-Controller interface

-Deadlock Detection Logic

CNC Lathe Machine

Tool wears detection and fault diagnosis station

-Accelerometer sensor

Quality analysis unit

-Eddy Current Inspection System

Central PLC Controller

-Failure detection

-Continuous Monitoring and annunciation

When the work piece is loaded into the system it will undergo four machining process, during this the shaft iron will be passed by conveyors, handled by robot arms, machined by CNC. Since the entire system is completely automated there should be feedback from every individual unit, even if one station running out of failure entire process will be disturbed. If it is not monitored properly entire work piece will be left as scrap so more care should be taken at each individual systems.

Adequate sensors providing sufficient information of the system are used. Once sensors come into picture suitable signal conditioning circuits are also taken into consideration. A central control unit is used in this case PLC, which receives all the sensory information in terms of voltage relating to the present status of the system.

To perform the machining operation CNC machine is used. The CNC lathe can machine pieces of soft materials such as plastic and wax, as well as harder materials such as aluminum, mild steel, and brass.

Our interest is to provide a machined work piece involving simple tooling operations. A CNC machine has the capability to do several operations which can be programmable through computer-aided design (CAD) and computer-aided manufacturing (CAM). For our purpose CNC is more than enough to do such turning, drilling, screw threading operations. Also such machines can be interfaced with the controller PLC.

Fig.3: Block diagram of various subsystems involved in FMS

Each sensor has certain capabilities and limitations, and thus the suitability of a given sensor depends largely on the application in which it is to be used. Some of the basic requirements are

small and rugged

reasonable energy consumption

operation and maintenance

selecting the measurement principle


able to function continuously and reliably


2.1 Conveyor Monitoring System

Belt Load measurement

To diagnose weather the correct raw soft iron work piece loaded in the conveyor system lets go for conveyor with load cell. The load cell determines the accurate weight of the work piece, which helps in getting the knowledge of loaded work piece.

Let the usual weight of raw soft iron work piece be x±0.25 kg.

Conveyor belt load cell should have high accuracy, after each machining the work piece weigh x tends to decrease since during machining the metal is wear out. The amount of metal removed will be approximately same for all work pieces. So on exact determination of the weight helps in identifying what are the machining processes completed on the work piece.

Raw work piece x±0.25 kg

After machining 1 (x-0.5)±0.1kg

After machining 2 (x-0.75)±0.1kg

After machining 3 (x-1)±0.1kg

After machining 4 (x-1.25)±0.1kg

These data's are fetched by the controller in respective digital levels which serves as feedback about the operations completed by both the machines.

Speed Transmitter

Based on the speed of the productivity all the individual systems should process coordinately. Productivity speed depends on the time taken to do single machining, material handling, speed of execution of the controller, robotic arm cycle time. So to have a coordinated process the belt speed should be regulated accordingly. For the measurement of speed of the conveyor speed transmitters are used. Also to match the time taken by the conveyor and machining operation same the speed of the conveyor drum should be monitored.


The belt conveyor scale a microprocessor based equipment specifically designed for industrial plant applications.

Main Features

Pulse generating speed transmitter

Opt isolated RS485 serial ports

Opt isolated 4-20 mA flow rate output

Failure alarm hardware output

Opt isolation of all digital inputs/outputs

Periodic self corrections of thermal drift

Load Receptor Specification

Nominal capacity 25 kg

Combined error ±0,1 % of nom. Capacity

Maximum load 150% of nom. Capacity

Operating temp. -20 +50 C

Speed Transmitter Specification

Power supply 24 V d.c. 20 mA

Nominal sensitivity 10 mm to steel

Operating temp. -20 +50 C

Protection IP65

Electronic Unit Specification

Power supply 24 V ,50/60 Hz, 15 VA

Load cell supply 11 V d.c., max 140 mA

Serial outputs RS485

Analog output 0/4-20 mA, 750 ohm max

A/D converter 16 bit "sigma-delta"

Nonlinearity 0,005 %

Fig.4: Belt conveyor scale "s4a"

Operating Principle

The belt load receptor supports one or more idler stations and delivers an electrical signal proportional to belt load. The instrument measures the output of each load cell allowing, thus an easy understanding of system problems such as incorrect load distribution among the load cells.

The speed transmitter is installed on a free running drum of the conveyor and delivers a pulse signal proportional to the drum speed and, thus, to the belt speed.

The electronic unit receives both signals and calculates flow rate and total quantity passed over the scale.

The Belt Load Receptor

The design guaranties an accurate belt load measurement independent from variations of the belt load center of gravity.

The Speed Transmitter

The speed transmitter model TVN1 is made by a high sensitivity proximity switch, which senses metal pieces mounted on a free running drum of the conveyor (tail or counterweight drum), thus guarantying no error measurement produced by slippage.

The transmitter generates one pulse per metal piece in such a way that the frequency of the signal is proportional to belt speed; this method guaranties zero error also in case of wide speed variations.

Real World Communication Ports

The opto isolated input/output board with four contact inputs, 1 open collector failure alarm output, 10 open collector outputs and one analog 0/4-20 mA output.

2.2 Tool Wear Detection And Fault Diagnosis:

It is very important to recognize the unmanned manufacturing system as an essential tool in manufacturing to avoid personal oversight and to enhance the productivity and quality of products. In order to insure efficiency within the system, the provision of monitoring facilities and algorithms for the adaptation of the manufacturing process should be executed accurately

Automated tool sharing systems provide a technological response to the high cost of tools in flexible manufacturing systems. These systems allow different machines to use the same tools by automatically transferring them from machine to machine as tooling needs evolve.

Tool Wears Detection And Fault Diagnosis Based On Cutting Force Monitoring

When a tooth enters and leaves the cutting material, it generates a cyclic cutting force from zero to maximum force and returns to zero for each tooth in one direction. The cyclic force looks like a peak. By expanding the principle to the resultant force generated on both x- and y- axes, and the number of peaks in each revolution is found to be equal to the number of teeth found on the milling tool.

Therefore, if the tool is in good condition, the peak force of each tooth should be the same for one revolution of the cutting process. Comparatively, if one tooth is broken, the broken tooth will generate a smaller peak force due to a smaller chip load causing the following tooth to obtain a larger peak force than normal.

Sensor chosen: Acceleration Transducer Mod. TA18S


The TA18S transducer picks up seismically the absolute vibrations of the machine by being fitted directly to the supports of the vibrating part it supplies an output signal directly proportional to the vibration of the point to which it is fastened. Such signal should subsequently be processed by one of the measuring channel.

Technical Characteristics

Type of measurement : seismic (absolute vibrations)

Dynamic range : ± 50 g

Frequency response : ± 3 dB 0,5 ÷ 15000 Hz

Direction of vibrations : any

Sensitivity : 100 mV/g

Power supply : 2÷20 mA - 18÷28 Vdc

Outlet impedance : < 150 ohm

Temperature range : -54°C ÷ +121°C

Connection : 2 pin connector MIL-C-5015 serie 3106/10

Fig.4: Diagnosis Based On Cutting Force Monitoring

2.3 Quality analysis unit

In order to inspect weather the machined piece has any surface flaws or near surface flaws either in symmetric or complex shaped the final automobile piece is subjected to analysis. If it passes the quality test it will be stacked at the output if else it will be separated at scrap piece.

SENSOR CHOSEN: Eddy Current Inspection System - EC2000

The EC2000 eddy current inspection system is a self-contained fully-automated eddy-current inspection system. The EC2000 is capable of detecting minute surface and near-surface flaws in either symmetric or complex shaped electrically conductive materials. The EC2000 improves both inspection reliability and productivity in comparison to manual inspections. It can be used for the analyze of round holes, bores, webs, fillets, dovetails, shaped holes, slots, scallops, spines, airflow path

Products Features

Multi-axis eddy current inspection system

CNC controlled, automated, 7 simultaneous-axes inspection of symmetric or complex shaped objects

Automatic data acquisition, analysis, storage, hard copy report and evaluation routines

Fully automated calibrations and inspection scanning


Instrument Type Flaw Detectors

Instrument Technology Eddy Current / Electromagnetic

Inspection Area Subsurface Crack Detection; Surface Cracks / Abrasion

Other Features Data Logging Capabilities; Programmable / Digital Control Unit

2.4 Proximity System

To detect the presence of work piece at each ends of the conveyor belt proximity switches are used. This information will tell the robotic arm that the work pieces waiting for arm to precede it. Also this information will be used to stop the conveyor belt once it reaches the end of the belt.

SENSOR CHOSEN: Inductive Proximity Sensor System P3001V10AI

Operating Principle

When a metal object moves into the inductive proximity sensor's field of detection, Eddy circuits build up in the metallic object, magnetically push back, and finally reduce the Inductive sensor's own oscillation field. The sensor's detection circuit monitors the oscillator's strength and triggers an output from the output circuitry when the oscillator becomes reduced to a sufficient level

Fig.5 working of inductive proximity sensor


Multi-voltage power supply24-240V AC/DC

2-wire-PLC compatible

LED showing output state

12mm detection range

P3001V10AI -Contact Normally Open

Fig.6 switch closes when the object comes near proximity


For some continuous, high-speed, high-volume processes, fixed automation remains the best solution is the use of industrial robots, which offer several advantages over fixed automation.

There are two basic types of assembly robots:

4-axis SCARA robots

6-axis articulated robots.

The acronym SCARA stands for 'selective compliance articulated robot arm'. This refers to the fact that a SCARA's arm segments, or links, are 'compliant', that is they can move freely, but only in a single geometrical plane.

The first two links of a SCARA swivel left and right in the horizontal plane. The third link consists of a metal rod called a quill, which holds the robot's end effectors, such as a gripper. The quill moves up and down in the vertical plane and rotates around its vertical axis, but cannot tilt at an angle.

This unique design gives 4-axis SCARAs a high degree of rigidity, which in turn allows them to move very fast and with high repeatability. In packaging applications, 4-axis SCARAs excel at high-speed pick-and-place and other materials handling tasks.

Fig.7: 4-axis SCARA robot

Six-axis articulated robots have two more joints than 4-axis SCARAs and, as a result, more freedom of movement.

The first link swivels in the horizontal plane like a SCARA, while the second two links move in the vertical plane. In addition, 6-axis articulated robots have a 'forearm' and two 'wrist' joints, which let them perform the same types of movements that a human forearm and wrist are capable of.

The additional joints of 6-axis articulated robots mean that they can pick up a part, no matter how it is oriented off the horizontal plane, then insert it into a package that may require a special angle of approach.

They can also perform many other operations Fig.8: 6-axis articulated robot

that might otherwise call for the dexterity of a human operator.

ROBOTIC ARM CHOSEN: 6-Axis Articulated Robot VS-6556em-P10

For our application its enough if the arm can able to orient in all direction with gripper mechanism as its end effectors to hold the work piece.


Model name VS-6556E

Max reach Approx.650mm

Max. Payload 5kg

Cycle time 0.49 - 0.59sec.

Position detection Absolute encoder

Drive motor and brake AC servomotors for all joints

Position repeatability ±0.02mm

Maximum inertia moment 0.295kgm2

2.6 Central PLC Controller

PLCs has the advantages of

• Cost effective for controlling complex systems.

• Flexible and can be reapplied to control other systems quickly and easily.

• Computational abilities allow more sophisticated control.

• Trouble shooting aids make programming easier and reduce downtime.

• Reliable components make these likely to operate for years

The PLCs were programmed with a technique that was based on relay logic wiring schematics. For simple programming the relay model of the PLC is sufficient. Reed switches are very similar to relays, except a permanent magnet is used instead of a wire coil. When the magnet is far away the switch is open, but when the magnet is brought near the switch is closed.

The various processes involved which should be taken care of FMS central controller are

Detection of correct work piece loaded at the conveyor

Read all the proximity switches and keep tracking of work piece

To command robotic arms to avoid ambiguity or deadlock

Conveyor belt load detection

Tool wear detection of the CNC machine

Continuous monitoring of all system for failure and annunciation

Machined work piece inspection for flaws


Why to select this controller?

For our application having five modules it's sufficient to select a PLC with 7 I/O ports. Also AC500 has four analog modules, alarm management and serial interface which quest our necessity.

Description of the hardware

Back-lighted LCD display and keypad

SD card slot

Plug-in communication modules (1 to max. 4)

Optionally with integrated Ethernet or ARCNET

FBP interface (for slave)

Two serial interfaces for programming

Expandable by up to seven local I/O modules


64kB memory, 2 RS-232/485 interfaces - programming, MODBUS/CS31

Communication Modules

Standard field bus system and integration into existing networks.

Ethernet coupler supporting the protocols TCP/IP, UDP/IP

The I/O modules

Digital and analog in different versions. Can be simply plugged onto the terminal units - for local expansion of the CPU (max. seven modules) and decentralized expansion via the FBP interface (max. seven modules, thereof max. four analog modules).


Serial or via Ethernet or ARCNET networks.

Engineering interface

Provides access from the programming system to an external project database in which the program source code of one or several automation projects is managed.


Five standardized programming languages: Function Block Diagram (FBD), Instruction List (IL),

Ladder Diagram (LD), Structured Text (ST), Sequential Function Chart (SFC)


For our automobile part it doesn't involve more complex works or complex machining. Any CNC machine can able to perform such operations of boring, cutting, drilling, grinding, milling, turning.

So any CNC Lathe machine capable of doing simple operations is sufficient to produce our product.

A model a with basic machining operation Sharp CNC SV-2412 Series


Precision hand scraping on all structural Components

Control- Fanuc 0i Mate MC

42 gallon roll out coolant tank

7.2" monochrome LCD display with tool path graphic

32 bit microprocessor

3 simultaneously controllable axes

PCMCIA Card Slot

Tool Length Measurement

Tool life management

Standard operation features

Keyboard type manual data input (MDI)

Input/output interface (RS232C)

Status display

Clock function

Current position display

Alarm display

Alarm history display

Help function

Actual cutting federate display

Automatic Tool Changer:

Model SV-2412: 10 station (armless type)



The entire system has several individual units. The controller should take care of each unit for their coordinated processing. The major problems and there troubleshooting are discussed below

3.1 Detection Of Wrong Work Piece

The conveyor belt will be loaded with a single work piece only. No two pieces will be there at the same time. When the first piece is taken for further processing there after only next piece will be loaded in the belt.

So the belt load receptor will give a proportional weight signal to the controller. The controller can able to distinguish between fresh raw piece, one machined, two machined, three machined or some other material other than soft iron work piece with its appropriate weight signal.

In case if the machine doesn't perform well there will be difference in weight between correctly machined work piece and a scrap piece. By training the controller with all these data's it can able to annunciate for wrong process.

Fig.9:Block Diagram of work piece detection

3.2 Deadlock Control Logic

Deadlock is the state when the robotic arm will be in confusion when two many operations assigned for the arm at the same time. So the arm should be provided with some set of logic in order to define such state of ambiguity.

Let's view the status of the each robotic arm when they need to handle loads it will bring us the knowledge when such ambiguity arises during continuous machining operation.

From our assumption

Initial time at the start of process 't'

Time taken by the conveyor 'T'

Duration of time to travel is (t to T)

Also to perform a single operation the machine taking time T.

The 1st work piece will be loaded at time 't', after the time duration of 2T 2nd work piece will be loaded. Likewise consecutive pieces will be loaded with the delay of 2T.

To illustrate the deadlock time duration it will be easy if we look at the sequence of operation.

Loading 1st work piece

Time: t-T T-2T 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R1 R2 R2 R3 R3 R2 R2 R3 R3 R1

U1 B1 M1 B2 M2 B3 M1 B2 M2 B4 Y1

Loading 2nd work piece

Time: 2T-3T 3T-4T 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R2 R2 R3 R3 R2 R2

U1 B1 M1 B2 M2 B3 M1 B2

Loading 3rd work piece

Time: 4T-5T 5T-6T 6T-7T 7T-8T 8T-9T

R1 R2 R2 R3 R3 R1

U1 B1 M1 B2 M2 B3

Loading 4th work piece

Time: 6T-7T 7T-8T 8T-9T

R1 R2 R2

U1 B1 M1 B2

Problem Statement

From the above we note that

While loading 1st and 2nd work piece there is no clash for the robotic arm at a given time. For each time period it has the single job to perform so there is no ambiguity arises.

While loading 3rd work piece robotic arm R2 will be in trouble to do two operations at the same time period.

After the time interval 4T-5T 1st work piece will be waiting at B3 for R2 to handle at the same time 3rd work piece will be waiting at B1 for machining 1. So there existing the deadlock.

Control logic:

During machining 1st and 2nd work piece there won't be any deadlock and each new cycle begins at a delay of 2T. The successive 3rd cycle begins at 4T, and then the maximum machined work piece will be given more priority if ambiguity arises.

The proximity sensor is normally open(logic 0), when the work piece comes into contact it will be closed(logic 1)

Once the machine M1, M2 finishes its machining process it will intimate the controller through the status signal S1 and S2.

If S1 is logic 1 then machining process of M1 completed

If S2 is logic 1 then machining process of M2 completed

Here is the rough logic of the entire process.

Default logic

If(any appropriate proximity sensor active || machine status signal active)

Make appropriate arm to handle

Fn(monitor all sub systems status signal)

Activate alarm if failure detected

Fn(interpolate status led, display respectively)


With reference to the fig. 3

Deadlock for robotic arm R2 exists when

(P2 &&S1)==1



Deadlock for robotic arm R3 exists when

(P4 &&S2)==1

Conveyor B1

If (P1==1)

Check for weight of B1

If (not matching any of the work piece wts)

Alarm and display



Get the command for belt speed

Run motor B1

Stop motor B1

For dead lock 1

If (S1==1)

Command arm R2 to handle machined piece

If((P2 &&S1)==1)

Command arm R2 to handle machined piece

Handle the new work piece next


Read weights of B1 and B3

Give first priority to belt of low weight

Handle the next one


First priority to S1

Next to successive belt


Default logic

For dead lock 2

If((P4 &&S2)==1)

First priority to S2

Next to successive belt


Default logic

3.3 Diagnosis Of Failure Of Individual Sub System

From every sub system a status signal will be forwarded to the central controller. The controller will be displaying the status led and display units. If any things go wrong like failure of sensor, undefined state, whole sub system failure, any interruption in the middle, controller will read these information and appropriate action like annunciation can be taken through effective programming.

Logic of Central controller

If(emergency stop button inactive)

Fn(Read status signal of all subsystem)

Display led's

Check for smooth processing

If(Any failure detection)

Alarming and indication of failure

Fn(Read proximity switches)

Fn(Read belt weights)

If(non trained weights)

Alarm and display

Fn(Read speed of rotor drum)

Get user command

Regulate motor speed

Fn(Read the inspection system)


If scrap detected command R1 to separate it

Fn(Read tool wear accelerometer)

Analyze for max resultant force



Shut down

Will the assumptions taken on the system executed in practice?

We assumed that both the time taken by the conveyor belt to reach its end and the time taken by the machine to perform a single task are same as 'T'. This can be executed in practice by

We cannot able to change the time taken by the machine for a single task, but we can able to change the time taken by the conveyor by appropriately increasing or decreasing the speed of the conveyor belt.

So there the speed transmitter plays a role in giving a electric signal corresponds to speed.


Thus the flexible manufacturing system is automated with the set of sensors and control logic which can able to detect deadlock in the manufacturing process and failure in the respective subsystem.